Most wireless data networks must deal with the problem of an imperfect transmission channel, and thus must employ some form data integrity validation at the receiver, typically an error detection algorithm such as a checksum or Cyclic Redundancy Check (“CRC”) on data packets. When such algorithms detect a transmission error regarding data that must be transmitted error-free, the system must employ a form of Automatic Repeat Request (“ARQ”) functionality, which includes communicating a re-transmission request from the receiver of the corrupted data to the original sender. ARQ functionality typically includes sending an acknowledgement (“ACK”) or a negative acknowledgment (“NACK”) message with an address field that uniquely identifies a specific node or set of nodes to which the ACK or NACK is associated.
Data networks that employ star topologies, such as systems with base stations, access points or gateways, typically must deal with transmission resource constraints at the central node due to the one-to-many relationship between the central node and end nodes that is inherent in the star topology. In wireless networks, these constraints typically manifest in time, spectrum or power resources.
Circa 2009, the Internet was in a stage of its evolution in which the backbone (routers and servers) was connected to fringe nodes formed primarily by personal computers. At that time, Kevin Ashton (among others) looked ahead to the next stage in the Internet's evolution, which he described as the Internet of Things (“IoT”). In his article, “That ‘Internet of Things’ Thing,” RFID Journal, Jul. 22, 2009, he describes the circa-2009-Internet as almost wholly dependent upon human interaction, i.e., he asserts that nearly all of the data then available on the internet was generated by data-capture/data-creation chains of events each of which included human interaction, e.g., typing, pressing a record button, taking a digital picture, or scanning a bar code. In the evolution of the Internet, such dependence upon human interaction as a link in each chain of data-capture and/or data-generation is a bottleneck. To deal with the bottleneck, Ashton suggested adapting internet-connected computers by providing them with data-capture and/or data-generation capability, thereby eliminating human interaction from a substantial portion of the data-capture/data-creation chains of events.
In the context of the IoT, a thing can be a natural or man-made object to which is assigned a unique ID/address and which is configured with the ability to capture and/or create data and transfer that data over a network. Relative to the IoT, a thing can be, e.g., a person with a heart monitor implant, a farm animal with a biochip transponder, an automobile that has built-in sensors to alert the driver when tire pressure is low, field operation devices that assist fire-fighters in search and rescue, personal biometric monitors woven into clothing that interact with thermostat systems and lighting systems to control HVAC and illumination conditions in a room continuously and imperceptibly, a refrigerator that is “aware” of its suitably tagged contents that can both plan a variety of menus from the food actually present therein and warn users of stale or spoiled food, etc.
In the post-2009 evolution of the Internet towards the IoT, a segment that has experienced major growth is that of small, inexpensive, networked processing devices, distributed at all scales throughout everyday life. Of those, many are configured for everyday/commonplace purposes. For the IoT, the fringe nodes will be comprised substantially of such small devices.
Within the small-device segment, the sub-segment that has the greatest growth potential is embedded, low-power, wireless devices. Networks of such devices are described as comprising the Wireless Embedded Internet (“WET”), which is a subset of IoT. More particularly, the WET includes resource-limited embedded devices, which typically are battery powered, and which are typically connected to the Internet by low-power, low-bandwidth wireless networks (“LoWPANs”).
It was assumed that Moore's law would advance computing and communication capabilities so rapidly that soon any embedded device could implement IP protocols. Alas, this has not proven true for cheap, low-power microcontrollers and low-power wireless radio technologies. The vast majority of simple embedded devices still make use of 8-bit and 16-bit microcontrollers with very limited memory because they are low-power, small and cheap. Also, the physical trade-offs of wireless technology have resulted in short-range, low-power wireless radios which have limited data rates, frame sizes and duty cycles, e.g., as in the IEEE 802.15.4 standard.